Home  5  Books  5  GBEzine  5  News  5  HelpDesk  5  Register  5  GreenBuilding.co.uk
Not signed in (Sign In)


Green Building Bible, Fourth Edition
Green Building Bible, fourth edition (both books)
These two books are the perfect starting place to help you get to grips with one of the most vitally important aspects of our society - our homes and living environment.

Buy individually or both books together. Delivery is free!

widget @ surfing-waves.com

Vanilla 1.0.3 is a product of Lussumo. More Information: Documentation, Community Support.

Welcome to new Forum Visitors
Join the forum now and benefit from discussions with thousands of other green building fans and discounts on Green Building Press publications: Apply now.

    • CommentAuthorwookey
    • CommentTimeSep 13th 2009
    I failed to find anything on here already, so sorry if this is an old question.

    I am planning a single-storey extension between house (cavity brick) and garage (single-skin brick). Planning to build to AECB silver levels. Seems to me that there is a terrible thermal bridge where the new wall/roof joins the existing one. Does anyone have details for how to minimise this?

    The obvious thing is to cut out a vertical slot in the wall so that the insulation can be made continuous, but this may lose too much strength, especially on (2-storey) house? What about cutting out part of it (leave in every 4th brick?) or putting in wall-ties of some sort? This would be much harder for the roof connection as cutting out horizontal bits involves putting in lintels which is a) harder and b) may not help thermal performance much.

    Suggestions welcome, including telling me that it won't really make any difference and I should stop worrying about it.

    Plan here showing basic layout and junctions with house/garage. Extension bit is 300mm wide wall in top-left corner.
    • CommentAuthortony
    • CommentTimeSep 13th 2009
    try this - and good luck


    You are right up against it with this -- have you cavity filled the house? what are the new walls for the top left extension made of?

    If it was mine I would ask an engineer to design me a beam so that I could remove the outside skin of the existing house to above insulation level, this will make the extension bigger and get rid of the "in house winter cooling system" that you dont want any way. The garage wall should get to about the same U value as the new extension wall.

    need any more help?
    • CommentAuthorwookey
    • CommentTimeSep 13th 2009
    Yes, the cavity is rockwool filled. The new wall construction is still up for debate - I am dithering between timber-frame+fluff with say wood-fibre boards for a bit of thermal mass or block+PUR. Rainscreen will be brick to match house. Most of top wall (on plan, actually SEish) will be glass as it's supposed to be a sun-room.

    Planning to add 100mm PUR to inside of garage wall.

    The main house is getting more internal insulation along with detailed efforts to seal up all the holes around/between 1st floor joists to attempt to make reasonably vapour-tight layer, so eventually most of your 'winter cooling' system will have been disabled. It was certainly pretty draughty as supplied (wiring is mostly in cavity so lots of half-brick holes where it comes out).

    Removing half the cavity wall is a radical idea. Have to think about that. I was imagining something rather less invasive just at the corner-joins, but that does make some sense. Perhaps expensive for the gain
    • CommentAuthorsinnerboy
    • CommentTimeSep 13th 2009 edited
    To maintain continuity of the insulation layer - and thereby eliminate cold bridging - keep to the same method of construction . Your scope of works is too small , in my opinion , not to extend using the same construction spec as the existing house

    It is a simple matter to "tooth" the new leaves to the existing by simple localised demolition and rebuilding at the existing corners

    100 Rockwool in 100 cavity is better than 60 PUR + 40 Airspace in 100 Cavity . If you use an AAC block for the internal leaf you will get a U value of 0.29 . This is less than the designed u value of 0.27 of PUR + Airspace - but it is more realizable in practice . Partial fill insulation is rarely built correctly i.e. gaps are left between the boards together with gaps between the back of the boards and the inner leaf . Cold cavity air migrates behind the insulation to undermine its performance .

    I would knock the garage for 2 reasons

    1. To maintain a continuous insulation layer and detailing as described already
    2. 100 brick is pourous and absobtive . It will soak in rainwater . PUR will not allow any breathing to the internal The hidden face of the PUR boards will become sodden moulds will thrive .

    Would you consider extending the left hand side wall down to join to the rear-most garage wall for more internal space ?

    What kind of roof is going on the extension ? ( Need to know to consider cold bridge detail there )
    • CommentAuthorsinnerboy
    • CommentTimeSep 13th 2009 edited
    Actually whatever the roof spec - decide where the insulation layer is - and where that layer abuts the external wall
    cut and tooth in a course of these .


    This way you break cold bridging affect of the existing external leaf .

    Don't forget to include a course of these too ....

    • CommentAuthorwookey
    • CommentTimeSep 14th 2009
    Very helpful comments sinnerboy - thank you. You make a good point about using same construction for this small amount of new-building. The problem there is that existing construction is a rather feeble brick, 55mm cavity, brick, so most of the insulation value comes from the 100mm PUR on the inside. So I was hoping to do something rather better than 'existing'. Your point that things are made easier by not choosing something completely different makes sense though. I was planning to go rather better than 100mm rockwool . More like 200mm, although perhaps that is excessive given the compromises of attaching to existing structure. Can one easily go to 150mm rockwool in 150mm cavity?

    The back face of the garage is actuall 200mm solid brick because next-doors garage abuts (and their storeroom (smaller part of garage/store) of it have been made into their utility room, so now insulated). Seems to me that would make 100mm PUR would be OK here?

    One thing I have never really understood: if brick is porous then shouldn't water leave through outside face the same way it got in? Especially here in dry east anglia.

    Problem with extending side-wall down to garage to 'fill in corner' (if I understand your suggestion correctly) is that current 'corridor' gets absorbed into garage. And garage/workshop will be seriously full of 'stuff', so we get poor access through (or we end up with useless acute triangular bit).

    Current suggested roof plan is mostly flat (to avoid difficulties with triangles, and annoying neighbours who have small courtyard immediately to left of new party wall with sun coming from our side). Small transverse pitched roof connecting between house and garage. I am actually quite keen to find a way of getting more height inside by having another pitched section parallel to exisiting house wall, leaving smaller flat section from about half-way to party wall. This makes for a complex roof, which I guess costs money? But should look a lot better as well as having space inside for hot air in summer. (I suspect this needs a pic to make sense).

    Roof construction details not specified yet. I'm imagining warm roof: timber plus PUR/fibre to U 0.1ish Architect sugested 2 polcarb triple-sheet rooflights to U2.2 which si a) ugly as sin and b) thermally poor. I reckon it'll be light enough by glazing almost all of SE (top of plan) wall (and a velux at N end in pitched roof and/or glazed door). I'd really like to be able to model solar gain for different window options but have to work out how esp-r works, as the only Linux software available.

    BTW is it normal to get digital plans out of architects? So far I only have paper which is useless for posting on here, and for modifying and sending back.

    I had thought of using foamglass at the base of the wall, but not at the top join - neat idea. And more practical than tony's :-) The cavicloak would go on top of the foamglass or below?
    • CommentAuthorCWatters
    • CommentTimeSep 14th 2009 edited
    Posted By: wookey
    One thing I have never really understood: if brick is porous then shouldn't water leave through outside face the same way it got in? Especially here in dry east anglia.

    Yes generally it does evaporate off the outside but the rate depends on the amount of water and any wind.

    I've seen some old bricks that appeared to hold water longer than normal. Not sure why but perhaps they pick up a layer of greasy polution on them so water penetrating down throught the wall can't escape so easily? In particular I was offered a lot of reclaimed bricks from an "old mill" that looked great but the sample the rep had felt heavy like it was saturated and had slightly a waxy feel. It just gave you the shivers handling it.
    • CommentAuthorsinnerboy
    • CommentTimeSep 14th 2009 edited
    Writing from Sunny Dublin ...

    110 is max cavity width in Ireland unless you have a structural engineer to sign off for a wider cavity width . Don't know about England - your Building Control officer - or another poster- should advise

    IF - the same applies ( IRL/ENG )and if you don't have a structural engineer then
    50 composite board ( 37 PIR + 13 plasterboard ) on a 100/100/100 Rockwool filled cavity will get you tu U of 0.21
    AAC block internal leaf will take you to 0.19

    I have a concern about adding 100 PIR onto your existing 100/55/100 walls . Simple rule of thumb is to keep the internal insulation layer to no more than half the amount in the cavity - to avoid interstitial condensation . In other words the internal insulation may work to cause your internal leaf to drop in temp. so that condensation forms there . I would look to add no more than 37 CPB ( 25 PIR + 13 plasterboard ) .

    If your garage walls are not exposed and if your neighbour has insulated on his side - then it is not a heat losing element . And if totally weathered by roofing - rain penetration is not an issue . It is akin to an externally insulated wall - sort of . You may even ( subject to your Building Control officers consent ) be allowed to do an Ru adjusted calculation - see pages 17/18 here ( again this is Irish guidance )


    Are the two Garage walls towards the rear 100 brick + exposed to rain ? . If so I would knock + rebuild .

    0.1 U value for the roof is onerous . 400+ Rockwoll , 250+ PIR .... It is important to understand that when elements are connected that detailing is vital . You can create a risk of localised condensation where two elements are combined which are very mis mismatched in terms of U Value . The devil is very much in the detailing ( wall / roof connection for example )

    In Ireland Architects do not usually issue CAD files - only PDF or Paper prints . ( Too easy to get ripped off )

    The cavicloaks should be located so they drain out over the roof flashing - this will be above the perinsul blocks if you use them

    Where internal partitions break the line of internal dry lining - to walls which are insulated only by dry lining - dry line them too ( the internal partitions ) for min 1m into the building with 37 CPB ( 25 PIR + 13 plasterboard )
    • CommentTimeSep 20th 2009
    wookey, I've played a bit with ESP-r but didn't put enough effort in to get to be competent. There's definitely a learning curve. IIRC, it doesn't do lighting calculations itself but uses Radiance - http://radsite.lbl.gov/radiance/ - so it might be faster to use that directly if it's just a lighting model you want. Alternatively, PHPP can estimate annual insolation through windows at arbitrary angles if the windows are in its product list or you have the appropriate U and g values.

    On the topic of what's available on Linux, I've had good luck running THERM under the latest builds of WINE.

    Cheers, Dave
    • CommentAuthorwookey
    • CommentTimeOct 9th 2009 edited
    Sinnerboy said: I have a concern about adding 100 PIR onto your existing 100/55/100 walls . Simple rule of thumb is to keep the internal insulation layer to no more than half the amount in the cavity - to avoid interstitial condensation

    I agree that I am pushing the limits a little here, but I did do a condensation analysis and found that there is a condensation risk only on a relatively small numbers of days in the year. I am also going to extreme lengths to ensure airtightness so that warm wet air from inside is not passing into the wall, and finally this is East Anglia which really is very dry so the walls will have plenty of chance to dry out. I also intend to stash some humidity sensors in the walls to keep an eye on things.

    Sinnerboy said: In Ireland Architects do not usually issue CAD files - only PDF or Paper prints . ( Too easy to get ripped off )
    How do they get ripped off? Once they've done their drawings and been paid why can't I have a useful version of what I paid for as well as a paper one?
    • CommentAuthorsinnerboy
    • CommentTimeOct 9th 2009
    Clarification . Once fees are paid up - CAD files can issue if required .
    • CommentAuthorwookey
    • CommentTimeMar 21st 2010
    OK, resurrecting this topic as things are actually happening. Currently in planning.

    There is DWG, DXF & PDF here, along with approximate sketchup model and pngs:

    I'm going to have a lot of questions about details here as my architect doesn't seem terribly green-building-minded so I'm going to have to tell him everything I want other than 'bog standard'.

    Construction is masonry+cavity. I'm planning to make the cavity 150-200mm (and generally steal useful details from tonyshouse handy drawings). Teplo ties seem to be the answer for wide cavity ties.

    So first real question is: What do people use for lintels with wide cavities? There will be a lot of glass in one wall. Something will be needed to hold up the bricks above here. If it's tricky then maybe just use timber rather than blocks on the inside above the glass?

    There will need to be a chunky post at the corner in the mostly-windows wall to hold up a beam which will take all the pitch roof weight and might have to take all the flat roof weight too, depending which way the fall (and thus joists) run(s) (still under debate due to grumpy neighbours). Again sugestions for details of how to integrate that post between timber windows so it looks pleasing are welcome. I guess steel would be slimmest but nasty thermal bridge. Oak post?

    I have plenty more questions but let's start with that.
    • CommentAuthorDarylP
    • CommentTimeMar 25th 2010
    With regards to your first question, Catnic offer 'wide' lintels now, just a little longer lead time.
    I imagine other manufacturers offer the same. We have just had some for 150mm cavity, not a problem.

    Posted By: wookeySo first real question is: What do people use for lintels with wide cavities? There will be a lot of glass in one wall. Something will be needed to hold up the bricks above here. If it's tricky then maybe just use timber rather than blocks on the inside above the glass?

    A steel cavity lintel will cause a cold bridge between inner & outer leaf, they are difficult to insulate properly & they get expensive as you go beyond 150mm cavity.

    Better to use two separate lintels for inner & outer leaf. If you have stone/concrete/render outer leaf, you can use precast concrete lintels. If brick outer leaf then you can use a brick angle lintel or Kevington & other do brick faced precast concrete lintels. If you have exposed sections of timber frame then perhaps you could use a reclaimed piece of timber & make it into a feature?

    There are a number of options for closing the cavities. PVC cavity closers are not very Eco & are expensive for wide cavities. You could use rigid polyurethane insulation with a DPC roll, a plywood window box or a combination of the two.

    • CommentAuthortony
    • CommentTimeMar 27th 2010
    Funny thing that that is all just as true for any size cavity --- Why ever do we use single steel lintels --- beats me -- simply not thinking is the only answer I can think of.
    • CommentTimeMar 29th 2010
    Posted By: tonyWhy ever do we use single steel lintels
    Presumably, cost! And historically a lack of concern about thermal properties.
    • CommentTimeMar 29th 2010
    Posted By: wookeyThere will need to be a chunky post at the corner in the mostly-windows wall to hold up a beam which will take all the pitch roof weight and might have to take all the flat roof weight too, depending which way the fall (and thus joists) run(s) (still under debate due to grumpy neighbours). Again sugestions for details of how to integrate that post between timber windows so it looks pleasing are welcome. I guess steel would be slimmest but nasty thermal bridge. Oak post?

    Put the post clear of the wall? Either inside the room or outside the building. You still have a point penetration but that's probably easier to deal with.

    It's a very complicated roof structure. I guess there's a reason for that?
    • CommentAuthorwookey
    • CommentTimeMay 2nd 2010
    The reason for the complicated roof is trying not to block next-door's light too much (not that thery were grateful and complained anyway) whilst not having it looking like a plain flat-roof 'cheap as possible' extension from the back and front, combined with a triangular site.

    OK. I now have some drawings of build/junction details, and do of course have some queries for the team.

    First off. Airtight junction at flat roof joists, and angled joist hanging:

    Here is the basic construction. Wall is plaster, 100mm dense block, 200mm dritherm batts, 100mm brick. Roof is joists, 16mm ply, breather membrane, 150mm PUR, 20-80mm taper PUR, EPDM membrane.

    The joists can either run parallel to the boundary wall, or parallel to the central beam. That leaves some angled junctions at one side of the other - either onto the conrete blocks, or onto the central glulam beam. Which is easier/cheaper? Attaching to the beam is my guess (screw/bolt/clever fixing joint?), but I don't know if the engineer gets sniffy (and it will increase the loading a little). Pics of both options attached.

    Angled fixing onto masonry is awkward as it implies a very wide gap in masonry if cut-in to wall, or some kind of sideways joist-hanger - do they exist?

    Joist hangers avoid the need to puncture the airtight wall apart from where they are attached. Not sure how to detail that best (goop underneath? some kind of sock round back of hanger?). Perhaps best is the tonytray: http://tonyshouse.info/tonytray.pdf but that detail is assuming a first floor. How best to connect to ply/breather layer in roof where there will be no plaster above?
    • CommentAuthorwookey
    • CommentTimeMay 2nd 2010 edited
    Hmm. Apparently only one attachment per post. So here's the elevation detail.

    Also online as png and dxf here:

    (Attachment seems not be working - probably due to resolution being too high - what are the limits? It's only 50K, but 3000x2000)

    Larger version of above joistlayout diagrm is:
    • CommentAuthortony
    • CommentTimeMay 2nd 2010
    A lot of tricky air sealing details to get to grips with and I was concerned about thermal bridging to the inner lintel?

    Looks like a nice extension though
    You could plaster the outer face of the inner block and make that the Airtightness line (you could do what you want inside this line and the joists could sit on the inner block as normal). There would be no need for a Tony tray, special socket boxes or loads of Airtightness tape. We have used this system in our last few projects and its a lot easier.

    The Radon barrier coming out under the wall is the Airtightness line, the plaster is taped externally to the radon barrier at the bottom of the wall. The airtightness line transfers to the inside through the ringbeam at the top of the wall. The OSB is taped to the ringbeam on the inside and the joints where the OSB boards meet are taped. The window frames are taped on the outside to the external plaster. This gives us an airtightness below 1 every time.
    • CommentAuthorwookey
    • CommentTimeMay 2nd 2010 edited
    Viking house. Interesting idea, but doesn't that require your brickies to be plasterers too? or to build the whole inner leaf, and joists/ringbeam then plaster it, before building outside leaf? How does the plastering over the ends of the joists work - there would be shrinkage and cracking, would there not? I wasn't planning a ringbeam at the top of the wall -sounds excessive. And do you have worries about long-term maintainability as you can't get to the airtighness layer afterwards to make any repairs (e.g on window replacement or if later service additions break into the cavity?
    Hi Wookey

    Some mesh in the plaster will stop the cracking at the joist ends.
    Building the inner wall first is probably best practice now, you may consider using Aircrete blocks for the inner leaf.
    You can plaster the reveals of window opes to make it easier to replace windows.
    The top block can be plastered all round to make it as airtight as a ringbeam.
    You will be also plastering the wall on the inside but you won't have top be as careful trying to maintain airtightness.
    • CommentAuthorwookey
    • CommentTimeMay 3rd 2010 edited
    So the idea here is that plastering twice is cheaper/more reliable than doing it once but with fidlier airtighness detailing. OK. I think I'm sold on that. My architect is going to look at me funny when I suggest this :-)

    I thought aircrete blocks were hard to plaster to and prone to it cracking off? That was one reason I decided on dense ones for inner leaf.

    No-one has anything to say about acute angled joist attachments?

    Meanwhile on to my next dilemma: How to support outer skin lintels at corner.

    I'm thinking that it is best to move the main post at the corner inboard of the window line to reduce thermal bridging (although in fact the post is probably no worse than the windows unless I get thermally-broken frames which may well be beyond the budget on this project). Having it inboard means it is nicely positionned to hold up the inner-leaf lintels, which are the ones taking all the roof joist/rafter loads. So that's all good, but the outer leaf has some bricks (4 rows+wallplate)above the windows and there doesn't seem to be much to hold them up.

    The window frame is inset in the insulation so there is no suppport from that. I suppose a second post on the outside might be the best way to deal with it? I was trying to have a clean line of windowframe round the corner, as this is the bit we will be looking atfor the next 20 years.

    Maybe extending the main beam to the outer leaf is not so terrible? I sized it to approx 215x190 so it will be something of a chunky hole through the insulation, whch I was trying to avoid.

    Maybe I need to go back to joined (both leaves) lintels rather than two separate ones so the inner leaf structure can provide support?

    I wonder if something clever with a custom steel arrangement could be done? I can't quite see it myself and expect its horribly expensive. There was originally a brick pillar at this corner but I thought it was ugly and took it out. I guess there needs to be _something_ structural here.

    Here is a plan view. There is an 'artists impression' of how it is supposed to look at http://wookware.org/extension/extension-singlepitch-low.png
    • CommentAuthortony
    • CommentTimeMay 3rd 2010
    I would be a little concerned about the longevity of plaster even on mesh. It will crack and also suffer from vibrations coming through the joists.

    Once done unmaintainable -- but I like less than one as a result.
    I've not seen angled joists built into a masonry wall. I would use a rim/perimeter board the full depth of the joists to the top of the inner leaf & fit angled joist hangers to that. Simpson Strong Tie & Cullen BP have angled joist hangers for most types of joist.

    I would use a Tony Tray under the rim/perimeter board sealed to the inner face of the inner leaf & plastered over. Fold it up the outer face of the rim board, wrap it over the deck & seal it to the membrane to the top of the deck.

    You could put a thin piece of EPS or some expanding foam between the joist and the plaster to stop movement/vibrations. We used a strip of External Insulation mesh and plaster in the vicinity of the joists, this needs to have a 60 year life span for Irish Agreament Certification. The plaster we used externally on the rest of the wall is 10 parts sand, 2 parts lime and 1 part cement so its flexible enough. What is the longevity of the tapes?

    If you are using this method tape the windows on the inside as well because water vapour can get behind the frames and condense where it meets the Colder external tape, so fill up around the window frames as usual with expanding foam and tape on the inside as well.

    I prefer joist hangers myself to sitting the timbers on the wall but I thought it might be easier for you to sit the joists on the wall because of the difficulty in finding angled joist hangers. Have you thought of fixing a timber to the wall and sitting your joists on that at an angle?

    There is no problem plastering Aircrete blocks you just have to wet them down well first and put on a scud coat.

    If you are insulating to a high level and achieving a high level of airtightness then any Cold Bridges like the steel you mentioned will become major Condensation areas. Better to keep steel either Cold or warm, why not Externally Insulate those two walls and glue brick slips onto the External Insulation.
    • CommentAuthorwookey
    • CommentTimeMay 3rd 2010
    David. Thanks for those suppliers names - very handy. But so far as I can see both only supply skew hangers with a minimum angle of 30 degrees. My skew angle is around 18 degrees. Very handy to find the glulam fixings though - I was wondering where to get them from.

    Simpson do some some neat hidden end-plate connectors in the glulam range. Maybe those can be used for beam/joist connections at tight angles and thus all the masonry ones can be simple perpendicular.

    VH, as you say, putting a timber along the wall and sitting the joists on that would work, but I suppose I was finding the idea of an extra chunk of wood below the ceiling line unappealing. I guess it's not really an issue.

    VH. ONe problem I've tought of with your 'plaster in cavity' concept is that the boundary wall already has a 6ft wall built on it so there is only ~400mm access after inner leaf is built. That's going to very hard to plaster in and impossible to build the outer leaf in so it'll have to be outer-leaf built first on that portion, which makes 'plaster-in-cavity' no-go. So I think I'm back to airtight plaster on the inside and a tony-tray round joist-ends.
    You could notch the joists and the wall plate timber so nothing is seen below the ceiling line.
    Can you use the boundry wall as the external wall and pump the cavity with EPS beads?
    • CommentAuthorwookey
    • CommentTimeMay 4th 2010 edited
    We can't use the boundary wall for 2 reasons. 1) the extension is 700mm higher than the wall 2) It's the neighbour's wall and on their side of the boundary so having an independent structure seems wise.

    The external insulation+brick slips idea is interesting. The bricks need to match existing building and it seems to me that the chances of them being available as slips is low, but perhaps any bricks can be made into slips? How do I find out?

    Actually perhaps it's a good idea as it reduces wallthickness a bit, improves insulation (it'd have to be rigid), and we don't actually have very much facing brick so the cost may not be a big deal. There would presumably be some bridging from a lot of relatively chunky fixings to hold 200mm of foam+battens+slips to the wall? Any good references to how it is done (I'm suppose to be sending detail design in today and this is a rather significant change).

    Like Tony I was trying to avoid anything too exotic to make life easy for the builders. Just holding up the window lintels off the insternal structure without too bad a bridge seems a more conventional solution. I did see some thermally-broken structural mounts at ecobuild which are no doubt horribly expensive.
The Ecobuilding Buzz
Site Map    |   Home    |   View Cart    |   Pressroom   |   Business   |   Links   

© Green Building Press